Mass spectrometry of protein complexes - from networks to structures

Mass spectrometry is well established in the fields of proteomics and metabolomics. Much less established is its role in deciphering the overall architecture of protein complexes. Using a procedure whereby we begin with the intact assembly we have shown that we can decompose the assembly into its simplest building blocks, look at their overall shape and then integrate them into 3D models of the overall complex. In order to do this we first have to validate the method using complexes of known structure. We have collaborated with a number of groups worldwide to obtain complexes with a variety of different topological arrangements and structures. Through a systematic survey of these complexes we hope to determine whether or not this approach will be suitable for use in deciding the overall architecture of unknown protein complexes. A further stream of our research involves isolating complexes at endogenous expression levels using a modified isolation protocol in which a single subunit is tagged and complexes are purified using affinity methods. We propose to continue and consolidate very exciting preliminary data that we have obtained for complexes involved in translation and splicing of RNA. The first of this series of complexes consists of the elongation initiation factors. By comparing and contrasting both human and yeast complexes we hope to gain novel insight into their subunit interaction maps and overall architecture. Similarly for the complexes involved in splicing, both yeast and human complexes will be studied, either in collaboration with Olga and E Makarov or by isolation from yeast strains grown in-house.

Recent developments have shown that we can extract, and maintain intact within a mass spectrometer, a number of protein complexes expressed at engoeneous levels. Through partial denaturation of these complexes we have shown that we can yield series of subcomplexes from which it is possible to construct complete interactions maps, 3D architectures and even atomic models. The challenge that remains is to turn this into a universal approach and to integrate it with restraints determined by ion mobility mass spectrometry. To this end we will carry out a series of validation experiments in which homomeric protein complexes of known X-ray structure will be decomposed into series of subcomplexes. Collision cross sections will be determined for these subcomplexes using ion mobility mass spectrometry. If we can establish conditions whereby this method leads to a method in which the overall topology and packing of subunits is retained in the subcomplexes, as in the intact complex, we will have a robust method for determining the overall architecture and packing of subunits within intact assemblies. In parallel with these validation experiments we will also continue to isolate complexes from their cellular environment, as close to the native state as possible. To do this we have adapted the TAPtag procedure and shown that we can retrain more transient /labile interactions than we were previously able to characterise. Our preliminary data for the elongation initiation factors and spliceosome complexes are particularly encouraging and contain a wealth of information about additional factors that are present within these complexes - often not considered in in vitro reconstitution experiments.